46 STARKEY DUNCAN, JR, Hyme. B, he Anthropology of Communication”, la F. E. X. Dan (0), aman Communication Theory (Nes York: Hot, Rinchat and Witton), 139, Jaffe J, and 8. Fedseln 1990" Rhythms of Dialogue (New York: Academie Pres JnhobuonR. 9Et_ Dissusion of “Factors and Forms of Aphasia” by A. R. Lavi, in A. VS. ‘Ge Reusk and M, O'Connor els), Ciba Foundation Sppostn on Disorders 29 Language (Boson: Lith, Brown), 10-82 Jakabson, Ry CG. M. Fant and M., Halle 1952. Preliminaries to Speech Antal: The Distinctive Feetarerand Their Correlates (Cambridge: MALT. Pres) Kendon, A. 1967 “Some Functions of Gaze-Dietion in Soil Interaction Acta Paycholoica 26,263. Kir, 1969" dn Approach to General Systems Theory (New York: Van Nostrand Rein: old, MeQuova, N-A. (2) I9TI The Natal History of ax Interview (= Mieraflm Collection of Mamacips Gm Cultaral Anthropology, Pieeth Series) (Chicago: The University of Chicago Jouph Regenten Libary Department of Photoduplicaton. Meluer, L., W-N. Moris and D. P. Hayes Wit “intreption Ovtcomes and Vocal Amplitude: Explorations fo Soci Paychophynie”, Journal of Personally and Soil Paychoegy 18, 392402 Scheeo,A:E- 1965" Stream and Strctre of Communicational Behavior (Philadelphia: Eastern Ponasivania Psychiatric Institut) Sebook, TA 1962 “Coding in the Evolution of Sigualing Behavior”, Bohevioal Seince 1963. “The Informational Model of Language: Analog and Digisl Coding in ‘Ablal and Human Communicaon® iP. L. Gaia (ed), NetwolLangucge ‘td the Computer (New York: McOraw-Hl), 47-6 ‘Tage, GL. 1958 -"Paraanguage: A Fist Approximation”, Stas Ox Linguistics 13,112 rage, G. Land HL. Sih, J 1957 i Dutie of Belish Sractare (Washington, D. C-: Amedcan Couns of {earned Societe). vans, V. 1970 "On Geng a Word ia Edgowis”, Papers from the Sixth Regional Mestng Chicago Linuise Socery (Chicago: Chicago Linguistic Society), 567-77, Starkey Dunsin isa asitant profesor of paychology atthe University of Chicago. His chetreearc interest fac facefteracon. Among bis eicipal pubiatons se: “Noaverbal Communication” (1969), and "Some Signals and Rules for Taking Speaking Tras in Conversations” (1972, GAIL JEFFERSON A Case of Precision Timing in Ordinary Conversation: Overlapped Tag-Positioned Address Terms in Closing Sequences i. A widely used transcription procedure is that of marking the precise point at which some ongoing utterance is overlapped by the talk of ‘another speaker. The version utilized here is to placo a double oblique Uf in the talk of an ongoing speaker to indicate the moment at which another speaker has started to talk. This procedure has provided en- counters with a variety of systematic phenomena, among which is the following. Aa ‘address term’ occurring in ‘tag’ positon (asa last particle of an otherwise possibly complete utterance)! is overlapped by the talk of the recipient ofthat utterance. So, for example [eDavyo] Dek: Is ita stretcher patient // Ma'am, Calle: 1s Uh yes be (8:10.68 6:3) Penay’ 0 skay. 1 see yu, ‘Agnes: AcMight / Dear, Peary: Bye bye, ‘Agnes: Bye bye, ‘That is, Desk's “Ma'am’ is overlapped by Caller's “It's” and Agnes" “Dear” is overlapped by Penny's “ 2 ‘The strutural-sequendal features of @ “possibly complete uteranss' have been siescribed at rea leagth by Harvoy Socks ins sees of unpublished lectures, For the purposes ofthis paper the crcl fearures involve that wil complete utterance nity ‘be deta with a sentence it ned not be, but may indeed consis ofa sole word land thatthe orcurrence of x posible complete uterance cas provide for a next speaker's stating total a the moment of completion,6 ‘GAIL ETFERSON ji This paper argues that the oocurrence of overlapped tag-positioned ‘address terms is not a trivial misplacement in startings and stoppings, tut an intensely organized phenomenon, involving a capacity for and an orientation to precise placement of talk on the part of conversationaiss ‘As an initial warrant for supposing that the phenomenon might yield to inquiry about its systematic features, two facts are adduced indicating at least thet participants could have grounds for seriously ittending to its opcurrence: (1) that these occurrences involve address terms, and 2) that address terms (and overlapped address terms) massively ‘occur in Closing Sequences.* ‘The literature on address terms describes them as loci for Formulating, ‘maintaining, and reformulating the status ofa relationship * The ocou rence of an overlap of an address term may be seen as having possible Serious consequences, since any overlapped term is potentially unheard, fand an “unheard? address term, given its relation-formulating properties, ray be particularly consequential. Talk that contains address terms is texpectably given careful attention by participants, WHEN it occur. If a place in conversation can be located in which address terms sySTIMATICALLY occur, sort of organization can be seen for them, and for participants” attention to them. One such position is atthe termination. ‘of a conversation, within Closing Sequences. And Closing Sequences are themselves a systematic locus of a variety of relational work, including the work done by address terms.4 This provides a specific, steucturally organized place for a participants orientation tothe occurrence of such 2 ‘Chsing Sequence’ Jess the sort of tak which presedes and leads up tthe tecmination of at ast, cephone conversations. Emmanuel Schell and Harvey Sick in "Opening up Closings, Senioea VI: 4 29-327, propose that a Closing Seyenoce ine dacrete conversational abet, whose workings invave partcnants 8 ovcular and eharstriable oentations aad tes. PP'Se for atample, Roger Brown and Marguerite Ford, "Address Terms in American Engl Jounal of Abmormal and Social Psycholory 62 (1961), 375-85, reprinted in Dittns (Langage Cals end Soety (New York: Harper and Row, 1969. DP Eing Ginan charecteriss the work of ‘osing salutations’ for somewhat uferarircastances thn the termination ofa single Ineraction in what may be a ‘Src of interactions On p. 87 of Ch. 3, “Supportive Tverchangss", Relais in ‘PaceNew Yerk: Base Books 971), he ik about closing statins insituations of fem aking He remark that soe apace maybe locus of rather more extravagant ‘pressions of esteem and aecion than such otersiractrally Sed postions as tesings, prossely bests participants speak in antkipaion of soon being Tes alate o ove another, Perhaps tis characterization may be adapied to tbe more {Saal creamatenes ofthe clesing moments of sage interaction in what may be a feces. of iterations, and lz here at support for the supposion that Closing ‘Sequences ml be sacral fa for eriou anton by participant tothe osurrense fod content of adres ters. PRECISION TIMING I ORDINARY CONVERSATION 0 things as address terms, and provides a warrant to look there for the resources available to deal with such a potentially problematic occurrence as an overlapped address term, and for systematic constraints as to selection among the available resources, which would thereby require jacticipants not merely to attend to and solve these problems, but to doso jinn orderly way, The task then is to discover how deeply and intensely organized such a phenomenon might be ii, A consideration of the orderliness of overlapped tag-positioned address terms will proceed as follows. A series of analytic resources will be developed, and applied to an analysis of an instance of ‘double overlap" Ime 2:14) Jean: ‘Wellgo ahead, en T'm sure she'll et fn touch with you about thet: Mel: Yah Jean; Otay, Mel: Mokeytoke. ean Thank you | / Mel, Mel: ‘Thank you || Jean. esa Bye |! bye, Mel: Bye. ‘The analytic resourees willbe developed along the following lines. () A recipient of some ongoing talk has the technical capacity to produce his talk with precision in relation to that ongoiag talk. That technica! capacity is a required feature of talk which abides by inter- action-based constraints upon the placement of a recipient's tal. (U1) Precision placement isa feature of conversation by whicha current speaker may position some objects within his utterance by reference to the recipient's capacity for and orientation to precision placement of his ‘responsive’ talk, The precise placement of a recipients tale can then ental alternative posible actions performed by the recipient and provided for in the design of the current speaker's utterance, specifically so that ‘any one of some set of actions has been appropriate, (Ill) The very appropriateness, collaboratively achieved, of one of those alternative actions (the one constituted by an overlap of some designedly placed object in the current speaker's utterance) itself poses problems, Participants have access to a varity of sotutions, and there are constraints upon which of those they ean successfully employ. The attempt by copatticipants to deal with the occurrence of an overiapped0 (Gan. IEFFERSON tag-positioned address term in an orderly fashion, using available and resttcted resources, accounts for the occurrence of the double overlap. 1. A first requirement for the analysis of overlapped address terms is to establish that conversationaists have a technical capacity to place theie talk with precision. If that can be showa, there are grounds for admitting the possibilty that overlapped address terms in tag position fare more than trivially misplaced stactings and stoppings. A variety of eurrences will be displayed in which recipients can be seen fo be starting up ‘immediately’. So, for example, ‘collaborative sentences” indicate that people can, with an entre sentence under way, find just the place to start up with @ ‘completion’ of a prior speaker's otherwise complete utterance. The following instances are from face-to-face interactions. {sh:29] Ben: ‘Aa‘there- there wa a least ten mites (of tefic bumper tah burmpe. Ethel: because a that, forsam Roger ‘Well yah I would've done something ike that maybe, but T wouldn't 'v made it a point. Al to bring it up, ‘And there are instances of recipients of some more or less protonged utterance coming i at just the proper moment with their own proposed ‘completion of the sentence in a prior speaker's as yet uncompleted vtterance. (6183:) Dan Well gebsyuh- ua you're sorra saying ‘then tht un that maybe even in the ease of Your father, his mother tis to keep hima //in kind of, Kea: “young. forsa:sa, Louise: [No a Sosbe is someone who || is a carbon copy of their frend, Roser: ~Atinks Pepsi. 1 Ameliborate esnical discussion of ‘olaboraies or Yin productions esrb ing the work ofa sacond speaker upon some peor speaker's tak may be found in ‘aon of Harvey Sacks’ unpublsbed lotus, staring wih series piven in 1967 PRECISION TIMING IN ORDINARY CONVERSATION 5. ors3:0) Keo: Instead my grandmother offering lim & Arn, of bee, she'll ay / "Would Lovie Wanna glassa milk?” ‘And there are instances of a recipient's attempting to achieve ‘saying. the same thing at the same time” as the ongoing speaker. [oTs:1:1:44 (ete, Louse i comparing the heights of Roger and Ken. Ken has side is ‘ix foot two and Roger hes said that he himself is five ten) Louise: ‘That’ B:ve ten en that's sx foot two? Now tall are you, Aly Roser: {flow cll'r you AL 3) Louise: Bh / heh beh! Al Tra five seven. (Roger talks very quickly, 20 that he and Louise both say "AT"simultaneoushy) [ers5:31 Da “The guy who doesn’t run the race doesn’t win bat doesn tose i Roger: byt tose it In the above fragment, Roger may have initially attempted to do the entire last phrase, “But 'e doesn't lose it” simultaneously with Dan. He produces a start on “But, sees where Dan is in the course of that phrase, cuts off, and manages to place “lose it” simultaneously with Das. ‘The foregoing series of occurrences indicate that a recipient of some sort of ongoing utterance has the technical capacity to select a precise spot to start his own talk ‘uo later’ than the exact appropriate moment to place itso that it wll sound lke a ‘continuation’ of the priorjongoing talk, An alternative notation, used on the ‘completing an uncompleted utterance’ may better display this. Dan: bis mother tiesto keep him = Ken: ou Dan: ina kind of, Lovie: [No a Sashe is someone who = Roper: itis Pes Louis: (ise carbon copy of thir fiend. Ken: she'll ay = Louise “Wanna glassa milk?" Kea: “Wouldu-”o ‘GAIL. JEFFERSON “That is, the ongoing speaker's ‘continuation’ and the recipient's ‘con- tiauation’ can be seen as simultaneously started phrases or clauses, with absolutely no gap between the former and latter. However, in thatthe recipiot/next speaker has ‘had time’ to see what is developing and work out the what and when of the ‘continuation’, the above ocurrences may not appear to bear upon situations in whic, after just one word which can stand as a complete utterance, a recipient begins to talk so that he overlaps an address term in tag postion to that single word. And such occurrences constitute the major portion of the interchanges being considered. It remains to be argued that such work is ‘within the technical capacity of conversationalists, Following presentation of a collection of overlapped address terms in tag position, consideration will be given to a resource to which coaver- sationalists may have aceets and by which they may be able to deploy their capacity for precise placement upon the occurrence of a single word. eo) Penelope: ‘Yeh alright | | Dear eanete Okey, Peoeiore: lye bye Jeanete: Bye bye end call —— Ladies 2:6:13) Ann ‘Yeah wel, things uh always work out for the || bes Bat: Oh, certainly Beth Alright | Ano. ee Un hah, Amn Okay, Beth Gye, Ana’ Goodnight, end call —— NB 9.10-68:6:3) Aagnes: ‘bh [if be down ‘a a few minutes? Penny O::P kay, 4) Peany Okay. I see yon. Ages: ‘Az Tright |] Dear, Penny: Bye bye, Agnes Bye bye, end cll i PRECISION TING IN ORDINARY CONVERSATION 3 ‘Thanks en awful lt (Ted Okay dear. (Talk Show a1av=114) Ted think enybody theh really looks ait, fand who n't personally hung up one way for another, has got tal be a litle bit fmbivalent about it Caller: ‘Ted Parsons? thank you very ms! “Te ‘Manic tokio tuh you || Sir. Caller: AVsight, bah, Ted! Good sight end call —~ Ladies 23-6) Lois ‘Okay well Iwill Well see you, — Rae Okay /] Dest. Lois: Bye bye Rae: Be, 10d call [Gem Matt: i see yuh t'morruh {go over hat other thing with you but / I'm not goin out sight. Phil Yeah. Phil: ‘Yeah well okeodoe = Mat: ve Pil: I jethoueht 'dask// yuh y'know ju /stincase Matt Matt Mat. Phil: Matt: Vil se yuh U'morrow up "tthe shop then Phil: = Yeh okay Buddy. Matt Okay || Phil Phil ory, Matt ‘Yeah! Tha | ns. Pil: Gye, Mat: Yeah. end call IFD:1:68) Mack Thank you, Jerry Mmm, Mack Oka3,54 ‘GAME sRFERSON Jerry: Right / ] Mack Mack: Byes, Jerry: Right. Mack: See yuh later, Jenny Okay, Mack: Okay, Jerry Okay, hh / hh Mack: See yuh coffee time, Jerry Okay. Mack: Riaright // Jer, sen: Right. end call IMc:t2:14) Jean: ‘We'll go ahead, en I'm sure she'll get in touch with you about the ti:me. Mel ‘Yah, Jean: 10%, Mel Okeydoke. Jean} ‘Thank you | / Mel, Mel ‘Thank you /{ Jean. Jean: Bye | / bye, Mel Bye, —~end cal ‘The materials displayed to argue the technical capacity of a recipient to start up ‘no later” than the precise moment at which a particular sort ‘of ‘continuation’ could be accomplished for an ongoing sentence involve ‘no more than the two utterances that constitute the phenomenon. If a siailar operation is performed on the overlapped tag-positioned address ‘terms, the claim would be that in many cases the recipient of a single word has heard it, recognized its work, ascertained what his appropriate response can/shoUld be, recognized a completed object, and has started to talk at the moment of perceived completion which virtually coincides ‘with its inception, ‘Such a claim need not be made. Instead, it may be necessary to isolate as relevant the entire sequence in Which the phenomenon occurs. It would thea be proposed that, as a NTENCE can provide a monitoring aid for a recipient, giving him a variety of usable information for the content and placement of his ‘continuation’, 0 2 SEQUENCE may similarly operat. For convenience, a sentence can be considered as a “horizontal packege" and a sequence as a ‘vertical package’, where perhaps specific sorts of sequences have theit components so arranged that they can be predictively ‘monitored and used in their course — in this case by appropriately | | PRECISION TIMING IN ORDINARY CONVERSATION 55 alternating users of alternating components of the package. Further, the ‘nest utterance” might be seen as analogous to the sorts of ‘continua- tons’ displayed earlier; that i, a recipient is to be seen as providing a perfectly placed continuation of the developing vertical package. ‘The fact that the collected overlapped tag-positioned address terms occur in Closing Sequences may now be brought to bear upon the claim of a recipient's capacity to place talk at the precise end of a possible ‘complete utterance consisting of a single word. Some detailed work has been done on a variety of sequence types, which argues that some of ‘them can be stringently formal and intensely organized in a varity of ways. A sequence is then not merely the name ofa series of utterances that happen to occur one afler another, but a type of orgenization that is possibly analogous tothe sentence, and that may provide for predictive mositoring by a recipient. So, for example, a participant in a Closing Sequence may, upon hearing “Okay”, not only hear it as a complete ‘lterance, but Know that at that point in the package he can or should provide, as 2 ‘continuing’ component, a second “Okay” — or, if the prior “Okay” was a second “Okay”, then he may know that a that point in the package he can or should provide “Take care” or “Thani you" or some situation-specifed continuing component, which will be appro: priately and expectably followed by a second ease of that class, to which he will return with a first "Goodbye", the recipient of which wil follow With a second "Goodbye" ‘Therefore, a recipient need not be described as dealing with some single word as it occurs, but can be monitoring where he is in the package now, in order to produce some appropriate next utterance/word/package component and find precisely where to placeit. That the prior component isa single word need not be problematic for an operation involving a predictively monitorable package. It should be noted that address terms, although they appropriately and expectably occur in a Closing Sequence, are not organized serially in the way that other components of that package are organized. An address term may occur before, within, or after, perhaps any of the serially organized Closing Sequence Components,” The convergence of an © Arango equence types bs been describe by various worker ia ams of thei structural organization. Fr example, Greeting Sequence, Gretings Bypass Sequenss, Closing Soqvenes, Side Sequences or Inrecion Sequences, bd such 'acuoo'seguences Announcement, Ofer, Invitation, aad Correction Soqnces. A feature of at least ‘ome of them shat conversational orient ote projected structrein ther conduct fot "base estos or ‘ontracted” and ‘expanded’ versions of some given sequence, 2 That adds terms are lacdin tag’ positions obvious engueh, Inthe coletion56 GAIL, JEFFERSON intensely organized Closing Sequence with the consequential but quite diferenly organized placement of address terms for any given Closing Sequence may account in part for the frequent occurrence of address term overlapping ‘The independently organized occurrence of Closing Sequence com- ponents and address terms not only provides systematically character- iaable problems for which participants may expectably have access (o solutions, but also, if this is a conceivable account forthe occurrence of overlapped address terms, supports the initial claim that this phenomenon is in the first place a product of organized behavior and not a more or less random misplacement. It wll tum out tha the sorts of placement events initially used to display a recipient's capacity for precise placement are quite gross compared (0 other sorts of placement work of which conversationalsts are capable. ‘A consideration of the more delicate operations will demonstrate a requited component of the argument being developed; that is, that the technical capacity for placement is met with an ORIENTATION to that capacity, such that it can be systematically and differentially deployed to accomplish a variety of coherent interactional activities, First of all, ‘starting up immediately upon a perceived reason to ‘talk’ isan event based on interaction, requiring an orientation to one's copartcipant, and an orientation to a variety of systematic constraints ‘upon one’s placement of talk. ‘The following fragments involve a recipient’s starting to talk prior ‘to completion of a WoRD within an ongoing utterance, It will be argued that such placement of the recipient’ talk is involved in coparticipant- oriented work. Of Cg Sans ning ead abs wm re an iste fa Sige ppt, aged Fanny thankyou vey me ‘ie hteung oem sa ay be areca sere as pacar) Nora: ‘Thank yu drling fer calling me. la: Alright be seeing yuh, Nora: Fast wanna be sue > Lil: Okay daring I'l tak uh ya. Now: Alii Lia: Bye bye, Now: Bye xd cal — PRECSION TIMING RS ORDINARY CONVERSATION 7 pcx) Cater Fire department, out atthe Fairview Food / mart the Desk: Yes. Desk: ‘We've already got the uh call on that ma‘am, (cba Gone ‘What's happening cay: Rilly not too much, — 1 jst ran up th'stal(uhs That's why Pm buf / fing en puting. Gane: on (FD aL :44-5%9) Caller: Downtown, though, she worked fer:: 1 dunno if you know Ruse Ople (| thorpe, Dok: Yeah, Iknow'm. Mm hay, Caller ‘She works fr him, Again, these overlaps might be viewed as somewhat random misplace rmenis of talk by the recipient; for example, that they are belated attempts at starting to talk immediately following the prior word. It will be pro- posed that there is some systematic work being done with the precise positioning of the recipient’ talk Capturing the recipient's work in the following way may help to make clear the sort of work going on. “Fairview Food!/YES.”, “That's why I'm hufl/OH:.” and *... Russ Ogle//YEAH,” Jf one were attempting to display “Yes 1 know who/what you're talking about” or “Oh T see what you're talking about” with generalized ackrowledgers like “Yes” or “Oh”, which do not particularly prove understanding, then the timing with which the acknowledger is produced may relate to its clarity and convincingness. It would be ill-advised to place a generalized acknowledger sooner than “Fairview FOOD" or “That's why I'm HUF ...” oF “Ruse OGLE ...”. There are simply too ‘many potential directions in which an utterance may move, given “Fairview” (hotel, dry cleaners, theater, etc}, “That's why I'm” [so tire, so excited, speaking so quickly, etc], and “Russ” [... ty Smith, cll Jones, Harper, Markevitch, eto}, for one to be assured of being conmct at that point and of clearly and convincingly showing he knows hae it correct. By waiting for the nest particle to be produced one drasti- cally reduces the odds against correctness. Considering that one may show that one knows what the other is talk about by using such techniques as completing the sentence of the8 GAIL JEFFERSON other's so far uncompleted utterance or by “saying the same thing at the sane time’, these precisely placed generalized acknowledgers may be seen to be doing another type of work; showing that while I know what you're talking about, I have specifically gathered that information from you as my informant now, and bring to bear upon it what T know independent. ‘Such an activity may operate for each of the prior fragments in the following way. For the reported fire, while providing « demonstration of relevant competence and knowledge, Desk specifically acknowledges Caller’ helpfulness by “recognizing” information she is delivering rather than eoproducing it, For the issue of “huffing and pufing”, Gene indicates that he noticed the phenomenon, was wondering about it, and that tbe problem has now been solved, He simultaneously deals with the pro- prities of noticing some possibly negative feature of his coparticipant; ‘hati, a notioer ought to mention the nice thing he has noticed before he is asked if he noticed it, and ought to leave it up to the one of whom itis 1 feature to mention if tis a possible action. In the Oglethorpe frag iment, Desk may be involved in selecting among a series of “knowables” including Caller, where that happens to be relevant to this particular interchange. Caller has attempted to call @ fireman known to him, has been put into contact with this one, for whom he claims acquaintance, about which Desk is unsure. It is possible that if one produced a "com- pletor’ ora ‘saying the same thing at the same time’ — Desk and Caller simultaneously saying “Oglethorpe” — that sort of thing might be the way Desk begins to claim, not that he knows Oglethorpe, but that he now realizes that he knows Caller’s wire by virtue of her being the lady who works for Russ Oglethorpe, and now Desk in some fashion knows Caller by virtue of his being the husband of the lady who works for Russ Oglethorpe. By using a generalized acknowledger, and placing it previsely after the occurrence of the first particle of “Oglethorpe”, [Desk claims that he does indeed independently know Oglethorpe, but that he only knows Oglethorpe, is recognizing that specific knowable by virtue of what Caller has so far said, and is not completing the inde- pendently known information as a preface to claiming recognition of other relevant knowables, as the “Yeah” is used as a preface to claiming “T know him” In general, the placement of these acknowledgment terms may be ‘operating under tWo fine constraints. One, @ ‘no later’ constraint, relates to displaying that one does have some independent information ‘and is using it, If one waits until Uae objct has been completely produced, PRECISION TIMING IN ORDINARY CONVERSATION 9 fone does not prove that one knows, one merely claims it, This ean be particularly important for accrediting a claim such as “We've already got the call on that” or “I know him", having preceded the claim with a generalized acknowledger. One accomplishes that by starting to talk before the entire word has been produced.* Perhaps one appears to risk having turned out to be wrong when the entire object is finally produced, and takes that risk on the basis of an assurance provided by Jndependent knowledge, where, as we have mentioned, waiting for the first particle of that object to be produced drastically reduces the risk ‘That may constitute one sort of no sooner’ constraint, that is, technically accomplishing a credible display of “I know what you're talking about”. A different sort of ‘no sooner’ constraint resides in displaying to the copartiipant that, ifT turn out to be correct, i isin part because T have received enough information from You to recognize what you are saying, where distinctive actions may thereby be accomplished. It may thea be that delicate interactional attentions are not only possible but requisite for such thing asthe display of "immediate recog nition’, One can (oF can fail to) acknowledge one's copacticipant, and that acknowledgment involves a ‘no sooner’ constraint on what might in the frst place be viewed as & recipient's having started to talk as soon as recognition occurred. One does not talk just as soon as recognition ‘occurs, but waits for that point at which various kinds of copartiipant attentions can be displayed. To convey an idea of the collaborative possibilities for such an opera- tion, end to suggest that not only need the recipient of some information be giving delicate attentions to the ongoing talk, but that che producer of that talk should orien to producingit with a variety of possible operations in mind, we briefly examine a joke thot went wrong. It takes place in a multiparty interaction, and shows that the possible systematicity of ‘sufficient information’ is operating for more than one recipient. 5 Sacks notes that recosmition, where tis soUoms by an ongoing speaker fom & seciplnt, 6 dage in sma fasion, the byproduct baing an ovedapped ongoing ‘terance. That i, one speaker, attempting to lst a signal of recognition frm his recipient, produces a sees of desciption componens, that series to be cit ito land abandoned at the moment the recipient recopies what being descibed. $0, for example, the following sor of interchange ‘A: Theard you wereat he beach yesterday. What's her aume, oh you know, the tll redhead tat ves across the sree from Larry? Te one wi drove im fo work the day hiv ear wae 1B: Oh Gina! ‘A: Yeah Gina She sald she saw you atthe beach yesterday.© can JerFERSON (oTs:2:6 Ken Ie be the tweny ninth. —(0.6)— would be / | Easter. Louise Not I's ne’ week: One more week left a” shoo J Roger: ‘Yanna hear somesbing funny, 1] [can’t read a calendar, Louise 'S the fourth week of March o) Louie: You | what? Roger ‘An’ I don’t know the months in order, an" ath pride mabsef with fa. hetbeh hn! 02) Louise: { (sing song) ) January February March, April May, June Ja — Roget liimow oc say "em T si Fea Louise fy, Augus’ September, December Novem Hoer Oe} tober. Ken: ities always // the A (sung) ) Thiety days hath September, Api Sune (and 0:2) ‘We focus attention upon Roger's overlap of Louise's recitation of the ‘months. Note that although Ken overlaps Louise's utterance, he does 50 ‘by starting up at the completion of Roger's utterance, which has ended before Louise completes her recitation. This may indicate that Ken is attending to such issues as utterance completion but has ‘heard enough’ of Louise's utterance to see, a8 Roger has seen, that it is going to be & straightforward recitation of the months in order. ‘Roger overlaps the recitation at a point precisely halfway through a ‘welveitem list the second half of which, that is, the seventh item, has been embarked upon, This may constitute both his decision and his copatticipant-oriented display that enough information has been delivered to suppose that Louise is simply going t0 recite all twelve months in order, and that he has reached that devision based on information provided by her. Louise's secipients may be using a known system for the production and reception of information, to which they suppose she is orienting as well, Given that one should orient one’s talk to what others know, they may understand that she, knowing that they are monitoring for what she is doing, has given them sufficient information by which to make a decision, Furthermore, insofar as they know this information, they may { PRECISION TIMING BS ORDINARY CONVERSATION hhave a warrant to interrupt to cUt off talk that is insensitive to what she sluould know they know. appears that in fat she did not intend thet the recitation be so treated She used a conventional procedure for constructing a joke, that is, ‘providing the ‘surprise’ at the end of something familiar. Insofar as this particular ‘something failiar’ is a welve-item list, and there nas been no introduction that might alert recipients to a forthcoming joke, by the time the lst has reached the halfway point her recipients may be permitted to treat what she is doing as a prolonged correct recitation and have ‘grounds so to treat it, and grounds to interrupt. She has, in effect, preempted some requisite orientation to the ‘suficient information’ system, with its ‘no sooner’ and ‘no later’ constraints with an orientation to a well-constructed joke Given these considerations of the technical capacity to statt talking fat the moment of a perceived reason to talk, and the copurticipant orientation that may variously constrain the precise placement of talk so that one may be seen actually to be waiting to start rather than as starting up immediately, interchanges which intilly appear to be the most routine sort of interaction, where a recipient begins to talk upon the COMPLETION of a prior utterance, will be examined as possibly involving “delayed” starts by the recipient, and investigated for the interactional bases of such delay. Following are instances of @ recipient's talking at the completion of a prior utterance, marking an utterance component that occurred ‘within it as something ke want(ed) to refer to. [Caught in passing] Steven: One, wo three, ( (pause ) four ive six, (pause) eleven eighe nse te, Susan: “Bleveo"? eight, nine, ten? ineiaio) ‘Agnes: I Perey goes with- Nixon I'd sure like that Portia: Who? ‘Agnes Percy. That young fella thet ub- his daughter was murdered. iNaa9} ‘Aan: 1 ida” get tuh vote I dectined tu state this time, whea T registred, so, Hust ub, didn't ‘get tah vote fer president so, Portia You delined- Whaddiyou mean. Agnes: Wall I vote Republican and Democrat.A COAIL JEFFERSON If permitting an utterance with a problematic component to go to completion is more than @ matter of simply abiding by a rule that pro- vides that a prior speaker ought to be allowed fo finish talking before a recipient starts (and the earlier materials indicate that recipients do, for a variety of systematic reasons and in an orderly fashion, star¢ up prior to someone else's completion), then such an event may be subjected to (questions about the interactional bases of what may now be seen as @ “delayed response’. ‘As-a technical matter, the recipient may reserve inquiry or comment under an expectation that, since the utterance is ongoing and not yet completed, further talk will address, remedy, modify, et. the problematic component, At a point where he can see that such action will not be taken, he may then provide his inquiry. This procedure may have inter- actional significance, That is, the occurrence of a delay may relate to some work specifically relevant to a situation engendered by the ovcurrence of « problematic component in someone's talk. In other words, the recipient of some problematic component may, by waiting for the utterance in which it has occurred to be recognizably completed by its speaker before marking that component as problematic, account for his marking of it and display coparticipant orientation. The recipient shows that he has permitted, and the speaker has not provided, an unsolicited self-correction (or clarification or modification) of tat problematic component. The recipient marks the problematic com: ponent W, BECAUSE the speaker has not dealt with it, and he did not mark it before because the speaker was expected to deat with it. The very display by the recipient of his having oriented to the speaker may then signal to the prior speaker that he did not sufficiently orient to his recipient, And it may be of some consequence which of a variety of correction-clarfication procedures is followed (eg, an unsolicited self= correction, a solicited self-correction, correction by a recipient at the ‘momeat of the problematic component's occurrence, correction by a recipient afler some intervening talk, etc). It begins to appear that events like starting to talk immediately and waiting are both selated to accomplishing a variety of interaction: based activites, THOSE activities characterized as ‘starting to talk im- mediately’ or ‘waiting’. That is, one may accomplish an ‘immediate start’ where accomplishment involves some waiting, And one may ac~ complish a ‘delayed response’, which turns out to mean starting to talk as soon as the utterance in Which a problematic component occurred is completed, The technical capacity to start up immediately and the PRECISION TIMING IN ORDINARY CONVERSATION 3 orientation to differentially placed starts are then resources snuck used. in the accomplishment of a variety of work, ‘The following fragment will be examined for the interplay of starting up immediately upon a perceived reason to talk and interaction-based delay, requiring an orientation to those issues by several participants, ‘Tracking some of the intricacies may serve to support the claims being made for the operation of immediate and delayed starts, ‘The materials are excerpted from a telephone call, Gene has called Cathy. Cathy’s thirteen year old son Rouald is, atleast at some point in ‘he call, inthe room with her as she is talking to Gene, As the conversa tion has proceeded so fur there is no mention of Ronald's Cathy. At one point Gene has inquired about Ronald. ence by {Goldberg 11:10) Cathy he's getting tbe quite a, big bo:y, he's -bh Ob I'd say he's about what ve thyee enna half = Arentehu Ronald, Five (ou. Five four en * weighs about a hunnerd'n thirdy ive pounds. = AAUUGGH! WHADDA- || (L-LIE)! Well Bow: — = Owright? How much d'you weigh ‘One awenty five. Ob one mrenny/ | five What's yu tyina make & fatty out'v'm? Focusing on: Cathy: cn 'e weighs about a hunnerd'a thindy five pounds. = Ronald: = AAUUGGH! WHADDA. L-LIE! Ronald’s utterance appears to be an instance of starting immediately ‘upon a percsived reason to talk. His outrage at the high estimate of his weight is placed at the instant of a possible completion of Cathy's utterance containing the estimate of his weight. It will be proposed that Cathy and Ronald are orienting to a series of issues related to the production of « problematic component and the placement of a response to it, which involve them in both immediacy and waiting ‘To develop this point, we isolate an object that recurs in this fragment ‘and is utilized at least once in its isolated state. We shall argue that it contains suficient information for its tecipient to grasp its import, and s0 recognizably contains that information that the recipient can legito (Gat. FFERSON mately start talking and can have acknowledged information received in just the way that “Russ Ogle...", “Fairview Food...” That's why I'm huf..” and “January February March April May June Ju..." contain recognizably sufficient information fer a recipient to legitimately start talking. The object will be called a“decad’and occurs in this fragment as “dhiry” and “twenty”, It is used by Gene as a vehicle for recognizably sulficient information. A rather elaborate consideration of that use will te developed and then turned to a consideration of the position of Cathy and Ronald vis-dvis “AAUUGGH! WHADDA I-LIE!” “Although there are three participants to this fragment, it seems to be the case that Cathy and Gene are officially and technically the ‘ones ‘in conversation” with each other. Technically, while Gene appar- cently can hear Ronald, Ronald cannot hear Gene; thus they are not in a postion to be in conversation with each other. Since Gene and Ronald are not in conversation with each other and Gene and Cathy ae, then ‘Gene's tak will be constrained by his orientation to Cathy as his oficial coconvetsationalst. This has specific consequences for the production of some of Gene’s talk, [At one point in the fragment Ronald produces the correction “One twenty-five”, Suppose that Gene has heard that correction, but in terms of his being ‘in conversation’ with Cathy, he has ‘overheard it, and to iknow it legitimately and use it for his own talk with Cathy he must ‘hear’ it from Cathy. He can accomplish this by starting to talk no sooner than the place where Cathy can have recognizably transmitted sufficient information; that is, by starting to talk upon the completion lof the ‘decad’ “twenny” (the frst estimate of Ronald’s weight having been “about a hunnerd’n thiedy-ive pounds”, the correction has been done when “one tvemny...” has been uttered). Such an operation might stand as a characterization and an account of that fragment Ronald ne oventy-ve, catty: (Oh one twenny | / Hive. Gene: ‘What'r yuh tryna make a fatty ou'v'm? “That the placement ofa recipient's talk within the talk ofthe ongoing speaker to accomplish norit familiaity with the information and ‘acknowledgment of a copartiipant is quite formal, may be demonstrated in the following way. It may be suspected that Gene's utterance was ‘generated by Ronald's talk and ‘merely’ placed at an appropriate point in Cathy's talk. This may, in fact, be signalled to a coparticipant. That is, Cathy may understand that Gene has heard Ronald, has come up with PRECISION TDAING IN ORDINARY CONVERSATION 65 something to say to Cathy about what he's heard Ronald say, ab is Aeferring to her right to speak after Ronald has spoken, both as proper liaison for Ronald’s talk, and asa structurally positioned next speaker in 1 Correction Sequence with Ronald. The relevance of this latter featuce ‘may then convey the information that Geng has been attending the entire sequence produced by Cathy and Ronss’. Depending upon what she will say, perhaps, he will or will not say the thing he came up with when he heard Ronald’s talk. He now signals both that he has heard enough of Cathy's utterance to determine what she is doing, and that that will not change what he had come up with in response to what Ronald said, It can then inform Cathy that Gene has been a witness to the interchange betweea her and Ronald, an interchange in which Cathy has necessarily produced her part knowing that Gene was attending to it, and perhaps orienting to his attending to it; that is, produicng her part of | it with Gene in mind, He may thus successfully orient to her as his proper coconversationalist and, by displaying that he was bringing to bear upon her talk information he already had available, indicates that he orients to her interchange with Ronald, as perhaps she has oriented to his witossing of it ‘The formal work is that of signalling both familiarity with the ongoing talk aN acknowledging the speaker of it, That formal work operates not only for utterances whose Very content gives away familiarity, but for ‘utterances whose content masks unfamiliarity. Whatever the content of ‘of the object, its placement at a precise point of ‘a0 sooner and no later" ‘within the talk of an ongoing speaker aevorplishes the display of familiar- ity and the display of copartcipant acknowledgment. “That the ‘decad! is a conventional ‘place’ for a recipient’ ‘no sooner, no later” work may be suggested by its occurrence in other materials. So, for example: (rio) stl: “There was a p'eecen’a there? with a great big log sin! —(0.6)— Heeadit in iz hand, C mean it was nota lid gan, it svat a: —(0.6)— ten, fie, tw any inch guint Seanete ‘Yeah, In this case the increment has been moving in fives, and “twunny” the last item ia a conventional three-part list. Starting to talk after “..aWu..c” Feanete signals both that she recognizes the series and is, for example, not expecting “twunny-three”, and that “twunay” will be the66 (OnIL seeTERSON last number Estelle will provide. Agaio, she displays her familiarity with and competence with such a device, and acknowledges her coparticipant. “That a ‘decad” can be a locus of such a variety of work warrants ‘examining less explicit events surrounding its occurrence. This will permit a consideration of some potentially delicate placement work by Ronald and Cathy. Suppose that Ronald, knowing that he weighs one hundred and twenty-five pounds, heaving “... a hunnerd’a thirdy...” (or even “a Irunnerd’n thir.."), understands right then and chere that the estimate is high by at least @ “decad’. Suppose he has the technical capacity (0 start talking at that moment. Why the delay? He may be attending to such issues as “waiting for utterance completion’, for example, to display that an unsolicited self-correction has been permitted by him and not provided by Cathy, which ean give a particular warrant to his expressed outrage; that is, it is not merely outrage at the occurrence of a high ‘weight estimate, but at its uncorrected status, given that he provided Cathy an opportunity to correct i ‘The ‘delayed’ comment is placed at the moment Cathy's estimate is completed. This may display some technical capacity for placement fon Ronald’s part. It may also involve a sort of collaboration between Cathy and Ronald, in which Cathy has produced an utterance that lends itself to just such treatment. That is, Cathy ean know that the weight component of her utterance is potentially problematic. It is specifically produced as an estimate, in an environment of an already corrected prior tstimate: “On I'd say he's about what five three enna half-Arentehs Ronald ... en 'e weighs about ..”. It will be argued that she performs two interactional tasks; that she (1) s0 constructs her utterance that Ronald ought to wait until itis completed to see whether and how he should respond, and (2) constructs it in such a Way that the problematic component turns out to occur just prior to completion of her utterance, which is specifically *completed” after the occurrence of the weight estimate, at which point Ronald, having been invited in as a hearer to this segment of the conversation with rights to correct, can and should stat talking. That is, she displays ‘that he should wait and displays that and when he should start talking ‘The utterance containing the weight estimate can be seen asa two-part list, Tt s constructed so that when it turns out to have been a two-part list, it is a corretly produced two-part list with its parts separated by “en! In its course, however, it might be developing three-part list, That part ofthe technique of coasructing at involvesdetalled recipietorenied PRECISION 1 or Which is a conventional list format. So, for example, ic cotld go: “Five four en e weighs about a hunnerd’n thirdy five en eez getting beter ooking every day.” The repeated "en" between each component of the Ist elthough it may not be requisite for a three-part ist, i certainly used, and specifically is ACCEPTADLE for a three-part lst and coanscr for a two-part list. This means that in its course, 2 hearer might find himself waiting to see when and how the lst will be completed. If tere isa thied component like “and getting better looking ...", that could modify the sense of the high weight estimate and perhaps oblige a diffe:ent response than “AAUUGGH! WHADDA L-LIE!" 'A brief consideration of the modification possibilities for an intended {and in this case aleeady embarked upon) response can be offered for the following interchange: Ronald ‘AAUUGGH! WHADDA- || L-LIE! Cathy Well how: = Catt’ = Owright? How much d'you weigh ‘There is a difference hetween “Well how much d’you weigh” and “AL right? How mach d’you weigh.” The latter is, roughly, @ challenge, It appears that a fist version of “How much d’you weigh”, started in response 0 the noise of a complaint, cut off and restarted after the ‘occurrence of... LIE!” is involved in acknowledging the overlapped word and modifying her response to fit the conTENT of the complaint in ‘contrast to its contextually (and acoustically) derived sense. That is, Cathy displays that her initial request for the correct weight was “too nild” considering what Ronald's utterance turned out to be; that the ‘modified, tougher stance is in RESPONSE to Ronald’s accusation, which ‘was not initially expected to take that form, “The modification possibilities ofa three-part list, as wel. as the poss bility of an unsolicited selfcorrection or a correction request (ct ‘Arentehts Ronald") provide grounds for Ronald’s waiting to see how the utterance will end, The fact that Cathy may know thst the weight cstimate is a problematic utterance component seems to lead her to ‘eaton to its jueturs is indicted by some work by Mishusl Canaban of the iversity of California at Irvine, Wo proposes tat varia. pronuciatios of SAND? within Het may serve systnaticaly to separate and combine possible Ist omponcts(uapabshed seminar pape) So, for example, in he tmepact it fe four EN 'e weighs e hungerd'N thy Sve pounds”, the cosas! between “EN” find N" speiealy mus that whatever folows “a huanerd™(vlch a possible Complete ist component, given that i ees ia an estimate envionment wish has Sot foul sock measurement tere as "eet", "ocho, el) part of hat Tit Component begun with "a hunnerd™ aad aot be earl 2s an upcoming hid st ‘omponent, folowing a competed second component ee _________.___._____@ CONIL JEFFERSON construct her utterance so that it will end upon the occurrence of that ‘component. This is not merely a technical matter but has some interac- tional import, For example, an embedded problematic component ‘may imply that the speaker does not intend it to be remarked upon, and supposes that reipient(s) will understand it or accept it, That may turn fut to be programmatic for recipients, who, for example, “let it pass’ taking it that the speaker has intended it to be passed. A problematic ‘component occurring at the end of an utterance may specifically pose « “test for recipients, At the end of the utterance a next speaker should talk, If he refers to the problematic component he may be responding to its accessible position as a signal to talk about it. Given its displayed accessibility, if he does not talk about it he may be seen as specifically claiming to understand it or accept it; that is, to have no need to talk bout it. Rovghly, the convergence of sentence end and problematic ‘component may signal the speaker’s awareness of the possible pro- blematic nature of that component, and may assign some specific work to recipient). In the materials st hand, it appears that Cathy has placed the pro- blematie componeat merely ‘more or less’ at the end of her weight estimate. That is, “.. a hunnerd’n thirdy five” is followed by more talk, *.. pounds.” The analytic result is an elaborate consideration of the constructing of the estimate, which peters out, It will be argued ‘that the placement of the problematic component can be technically characterized a “directly prior to a component which, by simply sig- nalling utterance completion, invites a next speaker to start talking’; that is, that “pounds” is strictly a transition-relevant device ina situation in which transition may need to be marked, insofar as the possibility of third lst componeat or sel-cozrection is available ‘A series of utterances in which the problematic component and sen= tence end converge will be brielly considered. (o1s2220) Roger: But the airs gotta come in dere an” the sie is sorta inftrated wi Hie ub piny dust, 6.) Kea: Does’ Bother me any, A any dust!? fors:s:371 Roger ‘An’ because he's seareda dancing he's goon {ano in private dl he learns how. PRECISION TIMING IN ORDINARY CONVERSATION 9 sim [And a gooilooking girl comes up 9 you and asks you, y'kno™m, Roger: (Git asks you to- Alright, [ors4:23) Jim: Like yesterday thre was 8 trace meet at Palisades, Rees Was thece. Isn't that ‘reform school? Recs? Roser: ‘Yeah. Kea veh Jim: ‘punch niggers an’ everything? Ken: ‘Yeah. Roger: ‘You mean Negro dontch Several observations may be made about these fragments.19 For the purposes of the analysis at hand, it will simply be noted that in the first fragment “piny dust” isthe end of the utterance, and in the other ‘wo, items like “Y'know,” and “an" everything?” end the utterance, ‘A way to formulate the three cases is that in each case soMertinc other than an immediate recipient response happens subsequent to the pro- blematic component, Since this is a discrete analysis inserted into a developing argument, it will be necessary to mention hete sometbing that belongs, and will be considered, later in the paper. For the later two fragments, the talk that follows the problematic components may be characterized as “utterance lengtheners', which indicate to the recipient ‘thatthe utterance can have been completedsothst hemay begin totalk, ‘while as well providing thatthe ongoing speaker has not stopped talking. “This may be seen as technique for specifically ‘avoiding’ just the sort of| cecurrence involved in the fist fragment; that is, a pause between the utterance containing the problematic component and the recipients" response Tt may therefore be a known feature of problematic components that ‘their occurrence generates a problem for recipients; i the speaker or is he not, himself going to talk about that object? If one does not intend 10 talk about it, one docs ot merely not talk. One properly signals that one does not intend to talk about it by REPLACING such talk witha signal that transition is now under way. It appears, thes, that there is a structure! place, ie, following a problematic component, in whieh the ongoing 1 For an eaborte consideration of thse sequences, see G, Jefferson, “Side Se- ‘quences in D, Sudnow (ed), Stud tn Soll Interaction (New Work" Free res, im.70 ‘Ont EFFERSON speaker ought to provide ErmuER talk about that abject oR a signal that he ddoes ot intend to talk about it. When the problematic component literally ends an utterance, a pause may result. Note that this pause need hot be characterized as a period ia which the recipients ponder suet things as whether or not the problematic component is problematic. That_may be decided as the object is occurring. The issue appears to be one of turn-taking proprieties In the situation of the weight estimate, “... pounds” occurs in the slot immediately following "... a hunnerd’a thirdy ive”. That its specifically related to transition work, in replacing tlk on the problematic component with talk indicating that no such talk will be done, is argued in the following way."... pounds." isnota necessary part of the weight estimate Equivalent objects, such as “feet” and “inches” do not occur in the height estimate, and are not missing from it. The word “pounds” is not required for widerstandinga number that has been introduced as “en e weighs ..”. Given that series has been established that does not use such terms, the jeeurrence of “pounds” for an item in that series, converging with a slot ‘that may specifically be connected with transition considerations follow- ing a problematic utterance component, may have its recognizable work specifically located there by its reipient(). I¢ may then be a matter of delieate coordination that Ronald does not start up immediately upon Rearing “thirdy ..", nor upon a possible sentence completion, “thirdy five ..., which, although it is a possible sentence completion, is perhaps quite obviously not a possible uterance ‘completion for an utterance containing a problematic component, and that he does start up immediately upon the completion of ".. pounds.” [Not merely may his outrage Deni from the structure Cathy has utilized for hee weight estimate, but its expression may be PERMITTED, in just the oom he uses, by her providing him such a signal as... pounds”, so that he need not be caught, as a recipient obviously can be caught, waiting {see what she will say next (where such an occurrence might provide for his expressed complaint to be done somewhat differently than it ean be done following “instantly” upon the error) ‘The foregoing consideration of the interplay of ‘immediate starting” and ‘delayed starting’ has yielded a detailed partial analysis of a single fragment of conversation, Various components of that analysis may handle other materials as they turn up. And it is hoped that in the course of that analysis some support has been gained for the argument that conversationalists have a technical capacity to place their tall with precision, and that that placement is both generated and constrained PRECISION TIMING IN ORDINARY CONVERSATION 1 by a variety of interaction-hased considerations to which pacticipants| are oriented, 1. With the technical capacity for and an orientation to interactionally based placement of talk as a resource, a second major aspect of the organized occurrence of overlapped address terms in tag position is ‘examined; that of DESIGNED PLACEMENT by an ongoing speaker. This was briefly considered in Section I, and will be elaborated specifically terms of the placement of address terms twill be argued that an ongoing speake: + stient to the possibility that a recipient may exercise his option to place his tlk so as to overlap 1 tag-positioned term or phrase. Iti evident that addess terms can: be designedly placed, in that there are systematic alternative loci for them in an utterance. They can go in variety of places, including ‘preface’ position and ‘tag’ position Obviously, preface-positioned address terms are not subject to overlap by a recipient, as are tag-positioned address terms. This would provide ‘grounds for the preferred placement of address terms in preface position if the mere occurrence of non-overlapped talk were an aim of some ongoing speaker. It seems to be the case that, in situations in which both Positions turn out to be used, tag position is used frst. And there may be seasons that tag position use is preferced. The following fragment is used merely to display the order of occur sence of positioned addtess terms for ‘similar’ utterances. Iv is excerpted from 9 face-to-face interaction in which two teenagers, in the presence ‘of a participating adult therapist, are disputing the distance of ‘Camp Glenoak’ from Los Angeles. Ken says it's 650 miles away and Louise says it’s “Not even 380 miles” away. Then: [oT 3:54) Kea: 1 know it's more than that, Louise, Louise Kiso", G0) Ken ( Ghouting) ) Louise, you're not there! 1 know i more than that heh Louise: Ws or, FFor this fragment it is merely noted that the address term “Louise” is associated with two versions of a same utterance; that its first occur renceisin ‘tag’ position and its second in ‘preface’; thet the first utterance hhas been ‘frustrated’ by its recipient, the second being a version of a ‘figst” responding to what has happened to the fist, and that there aren Gait. 1eeFERSON more than two parties involved, which provides that potentially any speaker has more than one possible recipient "Another fragment, affording alternative loci of address terms for same or similar utterances ina series, and also the possible preferential ordering, first ‘tag’ then ‘preface’, willbe subjected to more detailed consideration. ‘This is excerpted from a fuce-to-face interaction in @ work situation. Cassie and Carol work under Tracy's supervision. The fragment starts as Cassie is working and Carol enters. (s:b:22p5 Cassie: Hi Caro. Carel Wi, Er you doing that? 0) casi: Yor sp. That's what I'm doin’ Carl: How "bout if F spit it with 3 Case: Nagoh::! ato: 1.Gon" have (anything ese tuh do Cassie: Well neither do 1. = Tracy "You want sumpa to do Carol? = Casi: Uae tuh be here il five third azol you want something tuh do? Why, what, A fist consideration involves a possible operation of the address term in tag position for the specific utterance series in which it oocurs. The purpose is to convey an idea of the possibilty of designing talk by reference to the precise point at which a recipient could and might start talking, where different starting points ean constitute different actions by the reipient, whichever one occurs being appropriate, nonviolative, and provided for in the design of the prior utterance. We focus on Tracy's “You want sumpn to do Carol" and ask if thet ‘tag’ placement can be operating by reference to the situation in which it occurs, orienting to some relevant interactional issues and to some specific alternative actions a recipient might do. ‘With speci reference to this utterance series, although it may be quite ‘lear inthe overall context of the ongoing talk that Carol is the intended recipient of and proper next speaker to Tracy's remark, the immediately prior two utterances (Carol's “I don” have anything else tuh do” and Cassie's “Well neither do 1") provide @ potential ambiguity as to the addressee of “You want sumpa to do”. Provision of an address term somewhere in the question will clarify that ambiguity. The placement 1% Tramerption by Al “Touts, University of Califor, Ivins PRECISION TIMING 18 ORDINARY CONVERSATION 2 of the address term “Carol” as a ‘tag’ provides for alternative poss bilities: (1) If Carol respons immeciately to the question component of the utterance; that is, overlaps the occurrence of her name, then she has correctly heard the question «s addressed to her; the ambiguity was nonproblematic, and the occurrence of the tag-positioned “Carol” was unnecessary and unused. (2) If enough time elapses for Tracy to produce Carol’s name ‘in the clear’, then in terms of the potential ambiguity, the name has been appropriately produced, the speaker now having selected among the utterance’s possible recipients and having clarified who ought to respond. In effect, the placement of the address term in tag position cam EXPECT an immediate response but can HANDLE & delayed response; the delay" is legitimate, and the tag-positioned address term stands as solution to the cause of the delay ‘Whetiaer or not such attention is given by a speaker to a poteatiat ambiguity and its solution ie an utterance series that provides two conceivable recipients for a directed utterance, there is a rauch ‘generalise concerned in this particular situation to which the address ‘term in tag position may be altending. Roughly speaking, for an utter- ance that provides a task for its recipient, as does “You want sumpn tuh do?” in a work situation by a supervisor, issues of willingness and reluctance on the part ofthe recipient are relevant. A question that invites its recipient to do a task may be “well-constructed” if it orients to suck issuee; if, for example, it designs its components to make it possible 19 scrutinize the elapsed time between question and answer fora recipient's willingness or reluctance, not merely to speak, but to take up the task Where an address term used 10 clarify a potential ambiguity could conceivably go in preface position, then at the completion of “Caro) you want sumpa to do?” there i a potential pause. That potential pause would not be any pause after any uteerance, but a pause afer an Literance that offered a task, a pause which could signal not merely reluctance to speak, but reluctance to accept the task Tag-positioned address terms, however, can operate to add to the length of an ongoing utterance; they can be included among a series of utterables which provide that a speaker has not stopped talking although 2 possible complete utterance has been produced, where if he were to “have stopped’, an informative pause might occur. A whole series of vutterables can signal to a recipient that he may/should start talking ‘while providing thatthe current speaker has not stopped. For example:% COnIL errERSON Tabata} Gene: (Ok: : yeah it's justa buncha crap = Y'know? ~ Cathy, = en's, = Tuh, cathy ‘Wal i the money there though Gene tu compensate, you? Within Gene's utterance there are five points at which a recipient could legitimately stat talking, can be seen to have been invited to start talking. “The address term is one among a series of addable objects, atthe eomple~ tion of which a recipient may begin to speak. Not all utterance lengtheners are appropriate for specific situations; for example, things like "Y'know?" and “Uh” may be characterized as pleas, which a supervisor proposing a task to a subordinate ought not use, and things like "Huh", “Okay’", “Right?” may be characterized fs ‘demands’, which for a first offer may be inappropriate. An address term used in & multiparty situation may function as ‘merely a specitica- tion’, nether a plea nor a demand, while it provides a matrix in which the ‘currence of an immediate response is particularly marked and may be viewed asa display of recipients ‘willingness’, while the occurrence of a delayed response is particularly unmarked, in that the ongoing speaker js sill talking and hence the delayed response is not necessarily a display of ‘reluctance’. Such considerations ean account for the preferred placement of the address term inthe tag position, which supposes willingness and obscures reluctange, If this turns out to have been a “ist” occurrence, where the second is responsive to some abuse of the first, placement of the address term in preface position may be a warrantedly ‘less considerate’ version, potentially providing for a pause when the utterance is completed, pate which may be seen asthe intended recipient'ssilence and scrutinized for issues of reluctance For a tag-positioned address term to become a ‘mere sound’, = matrix upon which an immediate response to the utterance to which itis appended is displayed, is radically to shift its status from a key locus of relational work to & sound particle in the service of another (ype of interactional work. Further, it has so far been taken for granted that the occurrence of an overlap, an ‘interruption’, can be an appropriate conversational event, although a basic conversational rule that ‘not more than one speaker talks at a time" may imply that itis a violation.*® Sacks develos this ule and ts consequences for participants in a seis of un published ecues and in Ch. IC ofthe draft manaserptof pects of te Seen ‘Oreansain of Comestion, 10 be published by Prentice Hal One reason that overlapping talk may be a violation is that the over- lapped talk ‘s potentially rendered sequentially nonimplicative: it is ‘rested as unbeard, so thatthe developing course of talk may not include the overlapped talk, which was supposed to be included. It will be argued that there are specific vecasions of overlap that are appropriately nonconsequential, in which the overlapped object is “deleted” (rendered sequentially nonimplicative) by collaborative efforts on the part of speaker and recipi ‘The following metropolitan police department « fa clear instance of ‘deletion’ I is displayed as Radio, Hubble Uh send & ku ambulance to fiften ot four Garvey Steet. Kid hit by a cae Filteen oh four Gatvey | How bad iit ‘Yew sir that’s betwoen Belmont snd uh Tied fon Glenmore. Desk: Alright. Send a cat Caller: Send an ambiance too please Desk: Alright sir, end all: Caller has started talking no sooner and no later than the moment ‘of completion of Desk’s “Fifteen oh four Garvey" and has thereby turned ‘out to have overlapped "How bad is it", The overlapped question receives no answer or acknowledgment, and the question does not recur. This is a systematic ovcurrence, the product of the work of both parties. Caller’s utterance is specially neSPONSIVE to Desk's ("Yes sic"), REFERS (o it (ith “that’s”, which may relate to both Desk’s “Fifteen oh four Garvey” and Calle’s “.. fifteen oh four Garvey Street”; Le, refers to both in their relationship to each other), and immediately roLLOWs it. The very structure of Caller’s utterance, its components and its placement, ‘and its ‘timing’ can serve to explain why itis produced right here and now. It provides that Desk’s talk up to that point was entirely adequate in rs structure and components to elicit talk from Caller. It may theo signal to Desk that he NexD not talk further ‘At the moment of overlap, Desk can devide what to do about the further’ talk that he is in the course of producing. And he does have systematic options. IPhe intends to have his talk included inthe course of the developing sequence, and perhaps has recognized that Caller’s talk is appropriate and ought not be eut off, he can provide a frst com-